3974

has become a major pathogen of community and hospital infections on a global scale1,2. It can cau-se a variety of infections, including skin and soft tissue infections (SSTIs), pneumonia, endocar-ditis, bacteremia, and osteomyelitis. The mul-ti-drug resistance and high-extent resistance of MRSA bring many difficulties to infection con-trol. Therefore, health care costs have increased rapidly, and the mortality rate has also gradually increased. Moreover, the strain has a strong abili-ty to adapt to the environment factors and easily spread in the hospital. Historically, vancomycin has always been the drug of choice used to tre-at MRSA infections3. However, since the first staphylococcus aureus (VISA) with low-sensi-tivity to vancomycin was discovered in 1996, more and more VISA and vancomycin-resistant staphylococcus aureus (VRSA) have been repor-ted4. The status of vancomycin as gold standard to treat MRSA has been challenged. Therefore, some new antimicrobial agents are introduced into the treatment of MRSA infections, in which linezolid has obvious advantages5-8. It has excel-lent antibacterial activity in vitro and in vivo, and is less prone to the cross-resistance with other antimicrobial drugs of inhibiting protein synthesis. With the increasing use of linezolid, in the past ten years, a number of clinical trials have been conducted to compare the efficacy and safety of linezolid and vancomycin in the treatment of gram-positive bacterial infections (including MRSA infection)9-17. However, the re-sults are not exactly the same, which not come to a strong conclusion. Therefore, we collected these rigorously designed randomized controlled trials to perform a meta-analysis to systematical-

Abstract. – OBJECTIVE: This study aims to explore the treatment of methicillin-resistant staphylococcus aureus (MRSA) infection by us-ing meta-analysis method.

MATERIALS AND METHODS: Pubmed/Med-line, ScienceDirect, CNKI and Wanfang database were comprehensively searched to obtain the randomized controlled trials (RCTs) on linezol-id and vancomycin in the treatment of MRSA in-fections. We extracted features and information of included studies and selected appropriate ef-fect models based on the heterogeneity test re-sults. The funnel plot was used to analyze pub-lication bias.

RESULTS: A total of seven RCTs including 5376 cases met the inclusion criteria. Meta-anal-ysis showed that the clinical cure rate of linezolid group was higher than that of vancomycin group after treatment (OR = 1.85; 95% CI: 1.33-2.59, p<0.001) and follow-up (OR = 1.49; 95% CI: 1.17-1.91, p=0.001). In the microbiologically evaluable patients, end of therapy (EOT) MRSA clearance rate, and test of cure (TOC) MRSA clearance rate of linezolid were superior to those of vancomycin.

CONCLUSIONS: Based on the combined anal-ysis of randomized controlled trials, the efficacy of linezolid should be better than that of vanco-mycin in the treatment of infections caused by MRSA, but conclusions still need to be further validated by more well-designed RCTs of a large sample.

Key Words: Linezolid, Vancomycin, Meta-analysis, Methicillin-re-

sistant staphylococcus aureus.

Introduction

Since methicillin-resistant staphylococcus au-reus (MRSA) was firstly discovered in 1960, it

European Review for Medical and Pharmacological Sciences 2017; 21: 3974-3979

J. LI1, Q.-H. ZHAO2, K.-C. HUANG3, Z.-Q. LI3, L.-Y. ZHANG3, D.-Y. QIN3, F. PAN3, W.-X. HUANG3

1Dean’s Office, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China2Department of Nursing, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China3Chongqing Key Laboratory of Infectious Diseases and Parasitic Diseases, Infectious Disease Department, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China

Corresponding Author: Qing-Hua Zhao, MD; e-mail: leehaillan@yeah.net

Linezolid vs. vancomycin in treatment of methicillin-resistant staphylococcus aureus infections: a meta-analysis

Treatment of MRSA

3975

ly evaluate the efficacy and safety of linezolid and vancomycin in treating the infections cau-sed by MRSA.

Materials and Methods

Literature Search Strategy PubMed, Medline, ScienceDirect, CNKI and

Wanfang database were retrieved systematically from database building to September 2015. The search terms include “linezolid and vancomycin”, “MRSA”, “pneumonia”, “bacteraemia”, “skin and soft tissue infections”, and “clinic trial”. The pu-blished language and publication year were unli-mited.

Document Inclusion Criteria (1) Data were complete, prospective randomi-

zed controlled trials, one control group at least; (2) interventions were comparable, exposure to the same environment, in terms of linezolid and van-comycin in the treatment of bacterial infections, pathogens at least contained MRSA; (3) the ori-ginal documents have clear outcome variables (clinical cure rate, microbiological clearance rate, and the incidence of adverse reactions).

Document Exclusion Criteria (1) Experimental trials; (2) subjects were can-

cer patients or patients with neutropenia; (3) pai-red trials of linezolid or vancomycin with other antibiotics; (4) the trials focused on pharmacoki-netics or pharmacodynamics.

Quality Assessment Quality of each RCT was assessed by two

reviewers independently according to the qua-lity evaluation criteria in Cochrane systematic review’s Manual 5.0.2. Jadad score method was used to evaluate the methodological quality of the included studies, which was divided into 1-5 points (1-2: low quality study, 3-4: medium qua-lity study, 5: high-quality study). The evaluation content included whether randomized assignment and blinding method were used, whether the con-cealment method was appropriate, whether the quitting or lost cases were described. If the score was less than 2 points, it was removed.

Outcomes The outcome includes the following: (1) the

clinical cure rate of clinical evaluable patients (subjects with clinical manifestations that met the

inclusion criteria and exclusion criteria) after tre-atment and follow-up; (2) MRSA clearance rate after follow-up; (3) mortality after follow-up; (4) adverse reactions.

Data Extraction After reading the full text, data extraction was

performed by two reviewers independently ac-cording to uniform standards. The contents inclu-ded clinical characteristics (cases, sex ratio, and average age), intervention characteristics (inter-ventions, dose, and courses), and clinical results (cured cases, microbiological bacterial clearance cases, dead cases and cases of adverse reactions). Any disagreements about data extraction were re-solved by discussion.

Statistical Analysis RevMan 5.2 software (London, UK) was used

for statistical analysis; relative risk (RR) was used as the analysis statistics; 95% confidence interval (CI) was used to assess the efficacy and safety. Heterogeneity between trials was evaluated using I2; I2=0 indicated that various studies were ho-mogeneous; <25% indicated no heterogeneity, 25-50% indicated mild heterogeneity; 50-75% in-dicated moderate heterogeneity; >75% indicated relatively large heterogeneity; according to the Cochrane Handbook, I2<50% is generally consi-dered no heterogeneity between studies. If there was no heterogeneity, a fixed effects model was choosen; if there was heterogeneity, subgroup analysis and sensitivity analysis were performed to detect possible causes of clinical heterogenei-ty and statistical heterogeneity; if the results still had heterogeneity after excluding the interference of these factors, a random effects model was used for combined analysis. Publication bias was intu-itively reflected using the funnel plot: the distri-bution morphology of clinical data was analyzed to determine whether there is publication bias; if there was no publication bias, funnel plot was normal.

Results

Document Creening Results According to the search strategy, 88 relevant

RCTs were obtained. After excluding animal stu-dy, subjects <13 years old, paired trials with other antibiotics, a total of 32 documents were primarily screened. After cursory reading, 21 studies were excluded, including 7 duplicated studies, 7 studies

J. Li, Q.-H. Zhao, K.-C. Huang, Z.-Q. Li, L.-Y. Zhang, D.-Y. Qin, F. Pan, W.-X. Huang

3976

on hospitalized time and cost-benefit analysis and 7 pharmacokinetic or pharmacodynamics studies. In the remaining 11 documents, 4 literatures were further excluded since they had no MRSA results, with sample size <10 and with cancer patients as subjects. Finally, a total of seven valid documen-ts9-15 meeting the inclusion criteria were included. A total of 7 RCTs included a total of 5376 patien-ts. The characteristics of included studies were shown in Table I. The allocation concealment of 7 randomized controlled trials was inappropriate. Only two trials used double-blind control method and five trials specifically described the quitting and lost cases. Linezolid-treated patients recei-ved 600 mg of linezolid q12h orally or intrave-nously, and vancomycin treatment group received vancomycin 1 g or 15 mg/kg q12h intravenously. Antimicrobial therapy was performed for 7-28 d. There were no significant differences between the two groups in age.

Clinical Outcomes of Meta-AnalysisA total of four studies reported the clinical

cure rate of end of therapy (EOT) in clinically evaluable (CE) patients. It was better in linezolid group than that in vancomycin group (RR=1.85; 95% CI; 1.33-2.59; p=0.0002) (Figure 1A). A total of 7 studies reported the clinical cure rate of test of cure (TOC) in CE patients after follow-up. The result suggested that linezolid group was superior to vancomycin group (RR = 1.49; 95% CI; 1.17-1.91, p<0.001) in cure rate (Figure 1B).

MRSA Clearance rate of Meta-AnalysisA total of 3 studies reported MRSA clearance

rate. It was higher in linezolid group than in van-comycin group (RR = 5.23; 95% CI; 2.30-11.87; p<0.001, Figure 2A). A total of 7 studies reported MRSA clearance rate of microbe-evaluable TOC patients. The meta-analysis suggested that it was higher in linezolid group than in vancomycin group (RR = 1.55; 95% CI; 1.09-2.21; p=0.01, Figure 2B).

Publication bias analysis As shown in Figure 3, the inverted funnel plot

was symmetrical, suggesting the absence of pu-blication bias.

Discussion

In this study, a systematic review of 7 rando-mized controlled trials was performed to com-pare the efficacy and safety of linezolid, and Ta

ble

I. T

he c

hara

cter

istic

s of i

nclu

ded

stud

ies.

Th

erap

y

Q

ual

ity

score

sPu

blic

atio

ns

Stu

dy

typ

e N

um

ber

Pop

ula

tion

Li

nez

olid

gro

up

Van

com

ycin

gro

up

of

gro

up

Wilc

ox e

t al13

M

ulti-

cent

er, R

CT

363/

363

Susp

ecte

d re

late

d-M

RSA

In

trav

enou

s drip

: In

trav

enou

s drip

: 1g,

3

bl

ood

infe

ctio

n or

com

plex

skin

infe

ctio

n 6

00 m

g q1

2h, 7

-28d

q1

2h, 7

-28

d

Wun

deri

nk e

t al15

M

ulti-

cent

er, R

CT

74/7

2 Su

spec

ted

rela

ted-

MR

SA p

neum

onia

In

trav

enou

s drip

: 600

mg

In

trav

enou

s drip

: 1g,

4

q12h

, 7-1

4d

q12h

, 7-1

4 d

K

ohno

et a

l10

Mul

ti-ce

nter

, RC

T 10

0/51

D

efini

te o

r sus

pect

ed M

RSA

rela

ted

In

trav

enou

s drip

or O

ral:

In

trav

enou

s drip

: 4

pn

eum

onia

or i

nfec

tion

in so

ft tis

sue

60

0 m

g q1

2h, 7

-28d

1g

, q12

h, 7

-28

d

Wei

gelte

t et a

l12

Mul

ti-ce

nter

, RC

T 59

2/58

8

Susp

ecte

d M

RSA

rela

ted

infe

ctio

n

Intr

aven

ous d

rip: 6

00 m

g,

Intr

aven

ous d

rip:

4

in so

ft tis

sue

q1

2h, 7

-14d

1g

, q12

h, 7

-14

d W

unde

rink

et a

l14

Mul

ti-ce

nter

, 32

1/30

2 H

ospi

tal a

cqui

red

pneu

mon

ia

Intr

aven

ous d

rip:

Intr

aven

ous d

rip:

3

doub

le-b

lind

RCT

600

mg,

q12

h, 7

-21d

1g

, q12

h, 7

-21

d W

eige

ltet e

t al16

M

ulti-

cent

er, R

CT

66/6

9 Su

spec

ted

MR

SA re

late

d in

fect

ion

Intr

aven

ous d

rip o

r Ora

l:

Intr

aven

ous d

rip:

4

60

0 m

g q1

2h, 7

-14

d 1g

, q12

h, 7

-14

d St

even

s et a

l11

Mul

ti-ce

nter

, RC

T 24

0/22

0 Su

spec

ted

rela

ted-

MR

SA p

neum

onia

, In

trav

enou

s drip

: In

trav

enou

s drip

: 3

B

acte

rem

ia

600

mg,

q12

h;

1g,q

12h,

7-1

4 d

Ora

l: 60

0 m

g q1

2h, 7

-14

d

Treatment of MRSA

3977

Figure 1. Forest plot of effect comparison between linezolid group and vancomycin group in clinical cure rate. A, the clinical cure rate of EOT in CE patients; B, clinical cure rate of TOC in CE patients after follow-up. EOT: end of therapy, TOC: test of cure, CE: clinically evaluable.

Figure 2. Forest plot of effect comparison between linezolid group and vancomycin group in MRSA clearance rate. A: The MRSA clearance rate. B: MRSA clearance rate of microbe-evaluable TOC patients. TOC: test of cure, MRSA: methicillin-re-sistant staphylococcus aureus.

J. Li, Q.-H. Zhao, K.-C. Huang, Z.-Q. Li, L.-Y. Zhang, D.-Y. Qin, F. Pan, W.-X. Huang

3978

vancomycin in treating confirmed or suspected MRSA infections. The results showed that at the end of treatment and follow-up, the clinical ef-ficacy and microbiological efficacy of linezolid group were better than vancomycin group. Our study has several limitations. Firstly, research with inadequate allocation concealment should be excluded from the analysis. In this meta-a-nalysis, 7 studies did not mention allocation concealment method. Therefore, it may produce a selection bias. Secondly, the majority of in-cluded study did not mention the double-blind methods. Although the merger results were con-sistent with the original results, no blinded desi-gn may lead to overestimation. Therefore, there may be implementation bias in this study. Final-ly, there are some clinical heterogeneity among studies, which are likely to lead to different cli-nical outcomes and may affect the strength of meta-analysis and extrapolation of conclusions. In short, the meta-analysis showed that linezolid had better efficacy than vancomycinin for the treatment of MRSA-related infections. In addi-tion, linezolid had other advantages, including no need to adjust dose in patients with renal insufficiency, oral formulation with 100% bio-availability, and no need to perform therapeu-tic drug monitoring. Therefore, compared with the classic glycopeptide antibiotics, its clinical application will be more extensive. However, it still should be used reasonably to avoid accele-rating the generation and popularity of bacte-rial resistance. Also, the results may be limited by the flaws and potential bias of the analysis methods. Therefore, we should be cautious about the clinical significance of the conclusions of

this meta-analysis. Definitive conclusions still need to be verified by higher-quality and more rigorously-designed head-to-head randomized controlled trials.

Conclusions

Based on the combined analysis of randomized controlled trials, the efficacy of linezolid should be better than that of vancomycin in the treatment of infections caused by MRSA, but the conclu-sions still need to be further validated by more well-designed RCTs of large sample.

AcknowledgmentsThis study was granted by the Chongqing Medical Uni-versity Project (Grant No. XB0507) and National Key Clinical Nursing Project (Grant No. [2010] 305).

Conflict of interestThe authors declare no conflicts of interest.

References

1) BaBBi L, BarBanti-Brodano G, GasBarrini a, Boriani s. Iliac crest bone graft: a 23-years hystory of in-fection at donor site in certebral arthrodesis and a review of current bone substitutes. Eur Rev Med Pharmacol Sci 2016; 20: 4670-4676.

2) shenoy es, Paras ML, nouBary F, WaLensky rP, hoo-Per dC. Natural history of colonization with methi-cillin-resistant Staphylococcus aureus (MRSA)

Figure 3. Funnel plot of publication bias analysis. A, Clinical cure rate; B, MRSA clearance rate. MRSA: methicillin-resistant staphylococcus aureus.

Treatment of MRSA

3979

and vancomycin-resistant Enterococcus (VRE): a systematic review. BMC Infect Dis 2014; 14: 177.

3) ZhanG s, sun X, ChanG W, dai y, Ma X. Systema-tic review and meta-analysis of the epidemiology of vancomycin-intermediate and heterogeneous vancomycin-intermediate staphylococcus aureus isolates. PLoS One 2015; 10: e0136082.

4) MoraLes a, LaLueZa a, san Juan r, aGuado JM. Cloxa-cillin-susceptible Staphylococcus aureus with high MIC to glycopeptides. Ever we use cloxacillin? Rev Esp Quimioter 2015; Suppl 1: 25-29.

5) Watkins rr, LeMonoviCh tL, FiLe tM. An evidence-ba-sed review of linezolid for the treatment of methicil-lin-resistant Staphylococcus aureus (MRSA): pla-ce in therapy. Core Evid 2012; 7: 131-143.

6) eCkMann C, nathWani d, LaWson W, CorMan s, soLeM C, stePhens J, MaCahiLiG C, Li J, CharBonneau C, BaiL-Lon-PLot n, haider s. Comparison of vancomycin and linezolid in patients with peripheral vascular disease and/or diabetes in an observational European study of complicated skin and soft-tissue infections due to methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect 2015; Suppl 2: S33-39.

7) Peyrani P, raMireZ J. What is the best therapeutic approach to methicillin-resistant Staphylococcus aureus pneumonia? Curr Opin Infect Dis 2015; 28: 164-170.

8) yue J, donG Br, yanG M, Chen X, Wu t, Liu GJ. Line-zolid versus vancomycin for skin and soft tissue in-fections. Evid Based Child Health 2014; 9: 103-166.

9) itani kM, dryden Ms, BhattaCharyya h, kunkeL MJ, BaruCh aM, WeiGeLt Ja. Efficacy and safety of line-zolid versus vancomycin for the treatment of com-plicated skin and soft-tissue infections proven to be caused by methicillin-resistant Staphylococ-cus aureus. Am J Surg 2010; 199: 804-816.

10) kohno s, yaMaGuChi k, aiLkaWa n, suMiyaMa y, odaGiri s, aoki n, niki y, WatanaBe s, Furue M, ito t, Cross-da-Brera r, taCk kJ. Linezolid versus vancomycin for

the treatment of infections caused by methicillin-re-sistant staphylococcus aureus in Japan. J Antimi-crob Chemother 2007; 60: 1361-1369.

11) stevens dL, herr d, LaMPiris h, hunt JL, Batts dh, haFkin B. Linezolid versus vancomycin for the tre-atment of methicillin-resistant staphylococcus au-reus infections. Clin Infect Dis 2002; 34: 1481-1490.

12) WeiGeLt J, itani k, stevens d, Lau W, dryden M, kni-sCh C, Linezolid CSSTI Study Group. Linezolid versus vancomycin in treatment of complicated skin and soft tissue infections. Antimicrob Agents Chemother 2005; 49: 2260-2266.

13) WiLCoX Mh, taCk kJ, BouZa e, herr dL, ruF Br, iJ-ZerMan MM, Croos-daBrera rv, kunkeL MJ, knirsCh C. Complicated skin and skin-structure infections and catheter-related bloodstream infections: no-ninferiority of linezolid in a phase 3 study. Clin In-fect Dis 2009; 48: 203-212.

14) Wunderink rG, reLLo J, CaMMarata sk, Croos-daBrera rv, koLLeF Mh. Linezolid vs. vancomycin: analysis of two double-bind studies of patients with methi-cillin-resistant staphylococcus aureus nosocomial pneumonia. Chest 2003; 124: 1789-1797.

15) Wunderink rG, niederMan Ms, koLLeF Mh, shorr aF, kunkeL MJ, BaruCh a, MCGee Wt, reisMan a, Cha-stre J. Linezolid in methicillin-resistant Staphylo-coccus aureus nosocomial pneumonia: a rando-mized, controlled study. Clin Infect Dis 2012; 54: 621-629.

16) WeiGeLt J, kaaFarani hM, itani kM, sWanson rn. Li-nezolid eradicates MRSA better than vancomycin from surgical-site infections. Am J Surg 2004; 188: 760-766.

17) anGthonG C, kraJuBnGern P, tiyaPonGPattana W, Pon-GCharoen B, Pinsornsak P, taMMaChote n, kittisuPa-LuCk W. Effects of renal function on the intraosse-ous and inhibitory effect of prophylactic cefazolin in knee arthroplasty. Eur Rev Med Pharmacol Sci 2016; 20: 5216-5222.